Re: Not Picking Index

From: "George Pavlov" <gpavlov(at)mynewplace(dot)com>
To: "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, <pgsql-performance(at)postgresql(dot)org>
Subject: Re: Not Picking Index
Date: 2007-02-16 18:33:22
Message-ID: 8C5B026B51B6854CBE88121DBF097A8684DA43@ehost010-33.exch010.intermedia.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-performance

> Note the DROP INDEX will acquire exclusive lock on the table, so this
> might not be the greatest thing to do in a production environment.
> In PG 8.2 and up there is a sneakier way to do it that won't acquire
> any more lock than the statement-under-test does:
>
> begin;
> update pg_index set indisvalid = false
> where indexrelid = 'soandso'::regclass;
> explain analyze ...;
> rollback;

this really smacks of that four-letter word that starts with h... -- i
am glad we have finally come around on the subject :-)

seriously, this is a great technique and an enormous time saver during
query optimization. thanks for sharing!

george

In response to

Responses

Browse pgsql-performance by date

  From Date Subject
Next Message Mark Stosberg 2007-02-16 19:25:21 Re: reindex vs 'analyze'
Previous Message Steinar H. Gunderson 2007-02-16 18:32:04 Re: Not Picking Index